Energy Efficiency in Machine Tools - A Self-Learning Approach Giovanni Di Orio, Gonc ¸alo Cˆ andido, Jos´ e Barata CTS – UNINOVA, Dep. de Eng. Electrot´ ecnica, FCT Universidade Nova de Lisboa 2829-516 Caparica, Portugal {gido, gmc, jab}@uninova.pt Jos´ e Luiz Bittencourt Av. Oswaldo Cruz 112 / 903 22250-060 Rio de Janeiro, Brasil jlbitt@gmail.com Ralf Bonefeld Bosch Rexroth AG DE 97816 Lohr a. Main, Germany ralf.bonefeld@boschrexroth.de Abstract—Due to the growing demand to reduce the envi- ronmental impact, the manufacturing companies of today are encouraged to adopt new green methodologies, strategies and technologies for increasing the energy efficiency of their manu- facturing production lines. These solutions have a great impact on several productivity metrics including availability and costs. The continuous pursuit of productivity and particularly of machine availability has led to an increase of the total energy consumption in production plants. However, productivity gains can also be achieved by reducing the life-cycle costs of the manufacturing production systems. The research currently done under the scope of Self-Learning Production Systems (SLPS) tries to fill the gap between availability and efficiency by providing an innovative and integrated approach for ensuring the efficient utilization of the resources in machine tools. Index Terms—Machine Tool, Energy Efficiency, Data Mining, Context Awareness, Service Oriented Architecture I. I NTRODUCTION The continuous pursuit of productivity and particularly of machine availability has led to an increase of the total energy consumption in production plants. Moreover, the increasing demand for new, high quality and highly customized products has led to very flexible production machines able to quickly react to new production conditions. However, the desired flexibility is often achieved by oversizing machine components in the design phase, what leads to a decrease of the overall energy efficiency. Although production machines have become more efficient in terms of accuracy, cycle time and flexibility, there are yet some deficits, like the efficient handling with resources [1]–[3]. Machine tools correspond to a significant part of the electrical energy consumers in manufacturing plants, since their work is almost completely done by converting electrical into mechanical energy. After the decision of the European Commission in the frame of the Ecodesign Directive (Directive 2009/125/EC) to include machine tools in the list of products to be analyzed, the European organization of machine tool man- ufactures (CECIMO) designed a concept for self-regulation of the sector. This reaction was anyhow not only caused by the political decision, but also because of the fact that energy efficiency is gaining more and more importance in the market of production machines. II. BACKGROUND The energy consumption of machine tool is a function of the temporal power demand which is not static but, on the contrary, dynamic during the machining process. The typical power demand of a machining process is shown in Fig. 1 and provides a basis to recognize states and/or actions during the machining process. Fig. 1. Machine tool typical power profile [4]. The Fig. 1 confirms the idea that the power demand is basically constituted by a variable part and a constant part [1]. The variable and constant power represent together the minimal amount of power that is required to have the machine ready to run [5]. A more accurate distinction to the given power classification is also proposed in [4], where four power segments are considered, namely: • Fixed Power: is the power demand to guarantee the readiness of the machine tool. • Operational Power: is the power demand to operate the component during the machining phase. • Tool tip Power: is the power demand at tool tip to remove work piece material. • Unproductive Power: is the power converted into heat. As stated in [4], acting on fixed energy consumption is highly relevant for improving the energy efficiency of man- ufacturing processes throughout the different machines states,